A typical web browser receives data from a web server that defines the appearance and rudimentary behavior of a web page for display on a client system. In a typical scenario, a user specifies a Uniform Resource Locator (“URL”), which is a global address of a resource on the World Wide Web, to access a desired web site. Generally, the term “resource” refers to data or routines that can be accessed by a program. An example URL is “http://www.microsoft.com/ms.htm”. The first part of the example URL indicates a given protocol (i.e., “http”) to be used in the communication. The second part specifies the domain name (i.e., “www.microsoft.com”) where the resource is located. The third part specifies the resource (i.e., a file called “ms.htm”) within the domain. Accordingly, a browser generates an HTTP (HyperText Transport Protocol) request associated with the example URL to retrieve the data associated with ms.htm file within the www.microsoft.com domain. A web server hosting the www.microsoft.com site receives the HTTP request and returns the requested web page or resource in an HTTP response to the client system for display in the browser.
The “ms.htm” file of the example above includes static HTML (HyperText Markup Language) code. HTML is a plain-text authoring language used to create documents (e.g., web pages) on the World Wide Web. As such, an HTML file can be retrieved from a web server and displayed as a web page in a browser. Using HTML, a developer can, for example, specify formatted text, lists, forms, tables, hypertext links, inline images and sounds, and background graphics for display in the browser to present the rich graphical experience that users have come to expect while viewing information from the Internet. An HTML file, however, is a static file that does not inherently support dynamic generation of web page content.
If dynamic content is to be displayed, such as a change stock price or traffic information, a server-side application program is generally developed to handle the more complex client-server interaction. The server-side application program processes an HTTP request and generates the appropriate HTML code for transmission to the client in an HTTP response. An exemplary HTTP request may include parameters, such as data in a query string or data from web-based forms. As such, a server-side application program can process the parameters and dynamically generate HTML code in an HTTP response to the client. An exemplary server-side application program may dynamically generate documents containing appropriate HTML code using a sequence of one or more formatted text write operations to a memory structure. Thereafter, the resulting document is transmitted to a client system in an HTTP response, where it is displayed as a web page in the browser.
Developing a server-side application program can be a complex task requiring not only familiarity with normal HTML coding that is used to layout a web page, but also with programming basics, including one or more programming languages, (e.g., C++, Perl, Visual Basic, or Jscript). Web page designers, on the other hand, are frequently graphics designers and editors, who may lack programming experience. Furthermore, simplifying complex web page development can speed the development of new web content by any developer. Generally, development of a custom server-side application program also requires tremendous effort, so much, in fact, that developers are often disinclined to attempt it. It is desirable, therefore, to provide a development framework that allows a developer to dynamically create and process a web page with minimal programming.
One approach to minimize the programming requirements of dynamic web page generation has been the Active Server Page (ASP) framework, provided by Microsoft Corporation. An ASP resource typically includes Visual Basic or Jscript code, for example, to process an HTTP request that specifies the ASP resource as the desired resource and, thereafter, to generate the resulting HTML code in a HTTP response to the client. Furthermore, an ASP resource may reference pre-developed or third party client-side library components (e.g., client-side “ACTIVEX” controls) instead of requiring that the page developer write all components from scratch. However, in the current server-side application frameworks, the programming required to dynamically manage client-side user interface elements (e.g., text boxes, list boxes, buttons, hypertext links, images, sounds, etc.) within server-side applications can still require sophisticated programming skills and considerable effort. An unanswered problem exists in property encapsulating programming required to process user interface elements, so as to allow the web page developer to focus on other aspects of the web page.
Server-side processing of client-side user interface elements, however, may involve complex state management issues, which would typically fall to the server-side applications program. The state of a server-side processing module corresponding to the client-side user interface element represents the server-side module's properties and configuration. A stateless client/server model, however, dictates that in communications between client and server the server does not maintain the state of processes occurring between the client and the server, because the communications connection between the client and server may disappear unexpectedly.